Building and Hacking AI Agents and MCP Servers in the Real World

As AI agents and Model Context Protocol (MCP) integrations rapidly enter production environments, understanding their security implications is critical. This workshop bridges the gap between innovation and safety through live hands on of real-world vulnerabilities in MCP implementations and AI agents. Learn practical detection techniques for security engineers and red teamers and secure development patterns for engineers building with MCP and AI Agents.

This is a pure hands on fast paced workshop, planned to take the attendees through creating their own AI Agents, MCP Servers, identifying vulnerabilities in them and exploiting them for profit!

The workshop will have live labs, demos and working examples that attendees can try out on their systems to build AI components and see how real world vulnerabilities are discovered and exploited.